Risk assessment based on hierarchical fuzzy inference and prediction using kalman filter for underground facilities in smart cities

Abstract

Background/Objectives: In developed world, the trend of underground facilities construction is on the rise as this provide a way to maximum utilization of scarcest resource in smart cities i.e. space. However, these facilities can cause serious damage to property and human lives if not monitored properly. This paper presents a risk prediction and assessment mechanism using Kalman filter and hierarchical fuzzy inference for risk analysis of underground facilities. Methods/Statistical analysis: Sensors are installed at selected locations to collect input data by observing underground facilities. After necessary data processing, sensor data is used to compute risk index of individual factor that ultimately contribute toward total risk index of underground facility. For computing total risk index from given input factors, we use hierarchical fuzzy logic-based system by combining related components in a tree like structure. Experiments are conducted with four selected factors to estimate total risk index of underground facilities. Every risk factor is estimated from underground sensors data which may have noise and error. Noise and error from input data is removed using Kalman Filter algorithm. Findings: Results shows that without applying Kalman filter prediction, results for estimation of final risk index are not satisfactory. Kalman filter prediction helps in removal of noise in sensor reading and provides an effective way for the monitoring of underground facilities by forecasting critical issues in advance to avoid potential damage and can also assist in improving the maintenance work. Improvements/Applications: In this paper, we have used simulation data. In future, we will use real data to verify the proposed model for underground risk prediction and assessment.